Heat Capacities and Thermodynamic Properties of Ammonium Nitrate

Download as pdf or txt
Download as pdf or txt
You are on page 1of 12

August, 1967] Heat Capacities and Thermodynamic Properties of Ammonium Nitrate Crystal 1833

BULLETINOF THE CHEMICAL


SOCIETYOF JAPAN VOL. 40 1833-1844 (1967)

Heat Capacities and Thermodynamic Properties of Ammonium Nitrate


Crystal:Phase Transitions between Stable and Metastable Phases†1

MaSariOli NAGATANI†2 and Tetsuro SEIYAMA


Departmentof Applied Chemistry,Faculty of Engineering,KyushuUniversity,Hakozaki, Fukuoka

and Minoru SAKIYAMA, Hiroshi SUGA and Syuzo SEKI


Departmentof Chemistry,Faculty of Science,Osaka University, Toyonaka,Osaka

(Received March 22, 1967)


The heat capacities of ammonium nitrate were measured between 15 and 410°K for the

stable and metastable phases by use of adiabatic- and conduction-type calorimeters. The phase
transformations between stable phases; i. e., V-IV, IV-III, III-II and II-I, and the met-
astable one between IV and II, were all found to be of first-order type. The direct transition
between V and II, which is only realized for the sample treated with a minute amount of sur-
face active agent, was ascertained to be typical higher-order type. Heat and entropy of transi-
tion of five kinds of transitions were determined. These results conclude that the heat capacity
data by Stephenson et al. at low temperature region are much more reliable than the recent
values by Eichenauer et al. The possibility of the existence of the seventh phase below 156°K

has been proposed.

Since the discovery of anomalous peak in specific On the other hand, mainly Finnish workers
heat curve of ammonium chloride by Simon et al. performed systematic calorimetric studies of various
in 1927, there have been accumulated so many kinds of metal nitrate crystals which also exhibit
results with calorimetric investigations on various similar kind of phase transitions due to the disor-
kinds of ammonium salts. These informations dered orientation of nitrate ion. A review article
together with a variety of physical properties ob- on the relationship between the entropy of transi-
tained have been attracting many people to un- tion and crystal structure change on these com-
derstand this effect more in detail from theoretical pounds has been published quite recently by Newns
as well as experimental points of view. and Staveley.1)
The present material, ammonium nitrate com-
†1 Read before the 19th Annual Meeting of the
posed of these two, ammonium and nitrate ions, is
Chemical Society of Japan, Yokohama,1966.
†2 Present address:Department of Chemical
Engineering, Faculty of Engineering, Kagoshima 1) D. M. Newns and L. A. K. Staveley, Chem.
University, Kagoshima. Revs., 66, 267 (1966).
1834 M. NAGATANI, T. SEIYAMA, M. SAKIYAMA, H. SUGA and S. SEKI [Vol. 40, No. 8

TABLE 1. CRYSTAL STRUCTURES AND PHASE TRANSITIONS IN NH4NO3

still a very interesting substance to study in these ing considerably at higher temperature region from
respects and indeed demonstrates many phase the data by Stephenson et al.
transitions of rather complex nature. Already in In the present study we have measured the specific
1932, three papers were published independently, heat from 15 to 410°K with focusing our special

with regard to the crystal structure and thermal attention on the following few points: i) To re-
properties, by Hendricks, Posnjak and Kracek,2) examine the reliability of the recent data of specific
West,3) and by Grenshaw and Ritter.4) Since then, heat by Stephenson et al. and Eichenauer et al.,
these studies interested many workers from the ii) To check the possibility of presence of the seventh
point of above-mentioned order-disorder orienta- modification, iii) To establish the free energy rela-
tional rearrangement of ammonium and nitrate tionship between all conditions governing the direct
ions in the lattice. transformation from phase II to V which was first
Through the effort of these investigations, this suggested by Amoros et al.13) from structural view
crystal is known to exist in at least five solid phases point and later really found by one of the present
at ordinary pressures. These modifications with authors14) by treating the sample with minute
the pertinent data on crystal structure and transi- amount of surface active agent, and iv) To esta-
tion points are given in Table 1. blish the reliable numerical data of enthalpies
A sixth phase is reported to be stable above 442°K of each transitions.
at pressures higher than 9000 kg/cm2 by Bridgman.8) The last point concerns with the differential
There remains, however, still seemingly somewhat thermal analysis method in which almost all
contradicting or unsolved problems on a possibility workers check the availability of their instruments
of the existence of the seventh modification at lower with the detection of the phase transitions of this
temperature, suggested by spectroscopic, thermal or compound, notwithstanding that there is no com-
other studies by Crenshaw and Ritter,4) Jaffrey,9) plete set of their reliable data.
Vol'fkovich et al.,10) Theoret and Sandorfy11) and
others.
Experimental
As to the specific heat itself, it has been believed
that the precise measurement by Stephenson et al.5) Materials. A special, reagent-grade ammonium
established the conclusive data below 315°K. nitrate (Nakarai Chem. Ind. Ltd.) was recrystallized
Quite recently, however, Eichenauer and twice from ethyl alcohol (special-grade, Wak&Chem.
Liebscher12) published a new result which is deviat- Ind. Ltd.), and then dried under high vacuum (10-6
mmHg) for 10 hr. This sample was then powdered.
This drying and powdering procedures were repeated
2) S. B. Hendricks, E. Posnjak and P. C. Kracek. three times, one after the other, in order to remove
J. Am. Chem. Soc., 54, 2766 (1932).
C 3) completely the occluded mother liquor.
. W. West, ibid., 54, 2256 (1932). A specimen used for the measurement of the specific
4) J. L. Crenshaw and I. Ritter, Z. physik. Chem.,
B 16, 143 (1932). heat of metastable phase transition between phase V
5) C. C. Stephenson, D. R. Bentz and D. A. and II was prepared by adding a small amount of
Stevenson, J. Am. Chem. Soc., 77, 2161 (1955). additive to the original specimen, and this is referred
6) D. H. Everet and A. E. P. Watson, Trabajos to as"specimen-add'hereafter.† The additive used
reunion intern. reattiaidad solidos,3°, Madrid,1956,1, for the present work is octadecylamine acetate, the
301-319. amount of which is about 0.2 wt%. The additive
7) V. Hovi, J. Poyhonen and P. Paalassalo, Ann. was prepared by neutralization of octadecylamine
Acad. Sci. Fennicae, Ser. A VI, 42, 3 (1960).
8) P. W. Bridgman, Proc. Am. Acad. Arts Sci., 51, (Wako Chem. Ind. Ltd.) with acetic acid, according
605 (1916). to the ordinary method. The procedure taken for
9) J. Jaffrey, Compt. rend., 224, 1346 (1947).
10) S. I. Vol'fkovich, S. M. Rubinchik and V. M.
Kpzhin, Izvest. Akad. Nauk. S. S. S. R., Otdel. Khim. 13) J.L. Amoros Portoles and R. L. Banerjee,

Nauk., 1954, 209. Bol. R. Soc. Esp-Hist-Nat. (G),58, 165 (1965),

11) A. Theoret and C. Sandorfy, Can. J. Chem., 14) M.Nagatani, Kogyo Kagaku Zasshi(J. Chem, Sot.

42, 57 (1964). Japan, Ind. Chem. Sect.),68,424(1965);68,1842(1965).


12) W. Eichenauer and D. Liebscher, Z. Naturforsch., † See Ref. 5 for the effect of some additives on

20a, 160 (1965). the phase transitions.


August, 1967] Heat Capacities and Thermodynamic Properties of Ammonium Nitrate Crystal 1835

TABLE 2. SELECTED VALUES FOR HEAT CAPACITY OF NH4NO3


1836 M. NAGATANI, T. SEIYAMA, M. SAKIYAMA, H. SUGA and S. SEKI [Vol. 40, No. 8;

TABLE 2. (Continued)

addition was as follows: The additive was dissolved using the benzoic acid as a reference of heat capacity.
into benzene (special-grade, Wako Chem. Ind. Ltd.), The weight of samples employed for the conduction-
and pure ammonium nitrate was soaked into this solu- type calorimeter was 5.1536 g or 0.06438 mol for pure
tion. After vigorous stirring of the soaked specimen specimen and 4.8130 g or 0.06013 mol for specimen-
in the solution, the liquid benzene was evaporated com- add, respectively. The heating or cooling rate was.
0.2-0.3℃/min, except the transition regions. In
pletely at room temperature. For introducing the
specimen into the calorimeter cell a vacuum-type dry the operated temperature region, the experimental
box filled with dry nitrogen gas was employed for error of the conduction calorimeter is believed to be less,
avoiding the humidity from the sample. than±0.5%.

Apparatus and Method. The heat capacities of


stable phases between 15 and 320°K were measured
Experimental Results
with an automatically controlled low-temperature
adiabatic calorimeter, the details of which have been Heat Capacity. The results obtained are
reported previously.15) The weight of the sample used
was 18.4883 g in vacuo, or 0.23069 mol at a molecular
presented in three parts as follows; (1) the data by-
the adiabatic calorimeter for pure specimen, (2)
weight of 80.048.
The measurements between 275 and 410°K for pure
the data by the conduction-type calorimeter for
specimen and that between 275 and 340°K for specimen- pure specimen and (3) the data by the conduction-
add were the same in its main features as the one used type calorimeter for specimen-add.
hitherto in this laboratory.16) (1) The heat capacities observed by use of the
Standard substances used for the calibration of the adiabatic calorimeter are listed in Table 2 in
conduction-type calcorimeter were mainly benzoic
acid and potassium chloride for the higher temperature
region. This benzoic acid was obtained through the
courtesy of E. J. Prosen of the Chemistry Division of
the National Bureau of Standards as a Calorimetry
Conference Standard specimen. The potassium chloride
was obtained by recrystallization of the salt (Special-
grade, Wako Chem. Ind. Ltd.) from distilled water.
For the heat capacity of potassium chloride, the follow-
ing equation given by Ziegler and Messer17) was
adopted;
(1)
where, t is temperature in ℃. The data derived from

this equation are smoothly connected with those by


Berq and Morrison18) below 270°K which are believed

to be most reliable. In the present work, reliability


of Eq.(1)was confirmed between 275 and 389°K by

15) H. Suga and S. Seki, This Bulletin, 38, 1000 Fig. 1. The heat capacity of NH4NO3 between
170 and 260°K.
(1965). S16)
. Seki, M. Momotani, K. Nakatsu and T. ○ Series 3 ● Series 5 ● Series 6
Oshima, ibid., 28, 411 (1955). ▲ Stephenson et al.5)
17) W. T. Ziegler and C. E. Messer, J. Am. Chem. △ Eichenauer and Liebscher12)
Soc., 63, 2694 (1941).
Temperature scale at the top is for the upper
18) W. T. Berg and J. A. Morrison, Proc. Roy. Soc.
curves above the dotted line.
(London), A242, 467 (1957).
August, 1967] Heat Capacities and Thermodynamic Properties of Ammonium Nitrate Crystal 1837

chronological order. They are the ratios of the in- TABLE 3. SELECTED VALUES FOR HEAT
crease in enthalpy, ⊿H, to the rise in temperature, CAPACITY OF NH4NO3
⊿T. Each vaiue of Tav in the table is the ari-

thmetic mean of the initial and final temperatures


of the corresponding ⊿T. The ice point is taken

to be 273.15°K.

As is shown in Figs. 1 and 2 the heat capacities


reported by Stephenson et al.5) agree quite well
with the present results in the whole temperature
region with deviations less than 1%. In our meas-
urement, however, it has been recognized that
the observed heat capacities depend on the history
of the sample (Figs. 1 and 3). This fact will be
explained and discussed later.
While, the data given by Eichenauer and Lieb-
scher12)deviate much largely from ours, in particular
at higher temperature region of phase V, and fur-
thermore in phase IV.
We have found, however, that there is a very
small anomaly at 156°K which is clearly recognized

to be outside of our experimental error. This part


was measured repeatedly three times and the
results are shown graphically in Fig. 2 in exag-
gerated scale. As to this anomalous effect we
should like to refer later again.

Fig. 2. The heat capacity of NH4NO3. (the small


anomaly around 156°K).

○ Series 3 ◎ Scrics 4 ○ Series 5


▲ Stephenson et al.5)

△ Eichenauer and Liebscher12)

(2) The heat capacities measured for pure


specimen with the conduction-type calorimeter are
listed in Table 3. They are ⊿H/⊿T as mentioned

above, their approximate temperature ranges, ⊿T,

being shown in the last column.


These data are plotted in Fig. 3 with some results
by the adiabatic calorimeter.
Phase II below 357.25°K is of metastable state

and appeared only in the cooling measurement


after heating the sample up to phase I. The heat
capacities of phase II measured in cooling direc-
tion increase with decreasing temperature in the
region below about 350°K.

Validity of this tendency was ascertained by


getting the identical results with the repetition of
measurement in cooling and heating directions
between 325 and 350°K.
1838 M. NAGATANI, T. SEIYAMA, M. SAKIYAMA, H. SUGA and S. SEKI [Vol. 40, No. 8,

run in cooling direction (run 2), transitions II→

IV and II→V*take place at around 319°K and.

then the latter is followed by the V*→V transition*,

so the starting material for the run 3 is the mixture


of V and IV. In the run 5, the proportion of
transition IV→II showed an increasing tendency

compared to the run 3, but in these experiments,


the phase III which is most stable in the region,
between 305 and 357°K did not appear at all.

The polymorphic modification of the sample and


the assignments of the observed peaks of heat
capacities are also shown in Fig. 4.
The differential thermal analysis curves of transi-
tions V-V*-II which are free from transition
IV-II are depicted in Fig. 5 for reference. This.
Fig. 3. The heat capacity of NH4NO3. (Phases
IV, III, II and I)
○ From Table 3
○ From series 5 or 6 in Table 2

◎ From series 1,20r 3 in Table 2


▲ Stephenson et al5)
△ Eichenauer and Liebscher12)
Numbered arrows show the order and directions
of one of experimental series with the conduc-
tion-type calorimeter.

Fig. 5. Differential thermal analysis curves for,


transition V-V*-II in NH4NO3-add14).

heating curve is similar in its shape to the transi-


tions V-V*-II of runs 3 and 5 in Fig. 4. With
an assumption that the proportion of the-
V-V*-II part of the sample was kept constant
during the heating process (runs 3 and 5), heats
capacity of the each metastable phase was estimated
by the following calculation;

(2)
Where Cp (est.) stands for the heat capacity
of phase V, V* or II; Cp(run 1), Cp(run 3), the-
observed heat capacity in each run; and the value
0.430, the fraction which undergoes the transi-
tion V-V*-II in run 3. This value was de-
termined by comparison of the apparent heats
of transition IV-II in run I and 3, and equating its
Fig. 4. Metastable phase transitions IV-II and to [1-⊿H(run 3)/⊿H(run 1)], where ⊿H(run 1)
V-V*-II appeared in NH4NO3-add. and ⊿H(run 3)were 1860 and 1060(⊿H(run 5)=
Peak 1: V-V * Peak 2: V *-II
1270) J/mol, respectively.
Peak 3: IV-II
The heat capacities determined by this proce-
(3) The measurements with specimen-add were dure are listed in Table 4. The estimation with
repeated three times by heating and succesive the data in run 5 instead of those in run 3 is also
cooling procedures in the temperature range from possible, but this procedure was abandoned, since
275 to 340°K. All the features obtained by this
the results are rather uncertain due to the smaller
fraction of the transition V-V*-II part.
process of measurement near the phase transition
were reproduced in Fig. 4.
† The appearence of the metastable V*phase
In the first heating (run 1) of the virgin specimen, cited here were originally confirmed by DTA and X-
transition IV→II occurs and in the succcssive ray methods (see Ref. 14).
August, 1967] Heat Capacities and Thermodynamic Properties of Ammonium Nitrate Crystal 1839

Fig. 6. The heat capacity of NH4NO3.

○ Pure NH4NO3 ● NH4NO3-add


▲ Stephenson et al5) △ Eichenauer and Liebscher12)
× Crenshaw and Ritter4)

All the heat capacities experimentally determined


by the present work are presented graphically in
Fig. 6, together with the results by Eichenauer (3)
and Liebscher,12) Crenshaw and Ritter4) and Such an apparent continuity between the pure
Stephenson et al.5) We can see in this figure that sample and the specimen-add indicates that the
the estimated heat capacities for the V-V*-II minute amount of the surface active agent does not
process are smoothly fitted to the extrapolated affect the specific heat values, at least, within the
curves from the data around 260°K of phase V experimental error.
and from the values of metastable phase II around Temperatures, Enthalpies and Entropies
324°K(Table 3). Here, the heat capacity of phase of Transition. The observed temperatures,
Vin the range between 256 and 275°K was cal- enthalpies and entropies of the phase transition
culated by using the following equation; are summarized in Table 5. Transitions V-IV
1840 M. NAGATANI, T. SEIYAMA, M. SAKIYAMA, H. SUGA and S. SEKI [Vol. 40, No. 8

TABLE 4. SELECTED VALUES FOR HEAT TABLE 5. TEMPERATURES, ENTHALPIES AND ENTROPIES

CAPACITY OF NH4NO3-ADD OF TRANSITIONS OF NH4NO3

Values in parentheses are estimated by assuming


a normal heat capacity (Cp=3.98+0.6219T-
6.3357×10-4T2J/(mol・deg K))between 160 and

380°K.

and IV-III were so sluggish that their transition


temperatures were not so precisely determined, the
accuracies of their observed heats of transition
being decreased.
An upper limit for the transition temperature
of IV→II was determined by the measurements for

specimen-add, and a lower limit for II→IV was

obtained with pure specimen, since the transition


IV→II did not take place in pure specimen and in

the case of specimen-add transition II-IV was


almost forbidden around the true transition tem-
perature. The transition temperatures of V-V*
and V*-II were taken respectively as these tem-
peratures where the heat capacities reached the
maximum values. As is shown in Fig. 4, these
transitions are considered to be of higher-order,
since the shape of the heat capacity curve is of
characteristic λ-type with the maximum tempcra.

tures which are the same values in heating and


cooling directions with no hysteresis.
The values of Tt for the stable transitions in
Table 5 agreed with those in Table 1 within the
experimental error. However, the formers are
slightly less accurate, so, in case of need, the latters
may be adopted for the transition temperatures
of the stable transitions.
In addition, metastable phase II which appeared
also in pure specimen in the cooling curve trans-
formed to phase III just below the transition point
of IV-II, and consequently the heat of transition
II-VIII at 324°K was estimated as 590+10J/mol.

This is also listed in Table 5.

Thermodynamic Properties of
Ammonium Nitrate

The values of heat capacity, entropy, and also


of the enthalpy and Gibbs energy function of am-
monium nitrate crystal at selected temperatures
were listed in Tables 6, 7 and 8.
Table 6 is for the stable phases, and the values
August, 1967] Heat Capacities and Thermodynamic Properties of Ammonium Nitrate Crystal 1841

TABLE 6. THERMODYNAMIC PROPERTIES OF NH4NO3 60 and 320°K, and±0.5% above 320°K. The
(STABLE PHASE other thermodynamic quantities in the tables were
(in J/(mol・deg K)) derived from the smoothed heat capacity curves
and the heat of transition data of the stable tran-
sitions in Table 5.
Table 7 is for the metastable states of phase IV

TABLE 7. THERMODYNAMICPROPERTIESOF NH4NO3


(METASTABLE
PHASES,IV AND II)
(in J/(mol・deg K))

and II in the temperature range where phase III is


most stable. The thermodynamic properties of
phases IV and II in this table were determined by
continuous connection with the stable regions, re-
spectively.
Table 8 is for the metastable phases V, V* and
II.
In this case, the values of entropy, enthalpy and
Gibbs energy functions were calculated by numeri-
cal integrations with Eq. (2) and the estimated
heat capacity data.

of Cp° at round values of temperature were de-

termined by drawing the smoothed curves through


-the measured heat capacity data
. In this proce-
dure, however, the data in Series 5 and 6 of Table 2
were omitted for the reason mentioned later. The
extrapolation below 15 to 0°K was made by Eq.

(4)

Fig. 7. Enthalpy vs. temperature curves of


(4)
"Th NH4NO3.
e error of the smoothed heat capacity data is ---- Metastable phase V, V* or II in
estimated to be+1%below 60°K,+0.3%between NH4NO3-add.
1842 M. NAGATANI, T. SEIYAMA, M. SAKIYAMA, H. SUGA and S. SEKI [Vol. 40, No. 8

TABLE 8.THERMODYNAMIC PROPERTIESOF 0.032in J/(mol・degK), respectively. Owing to the

NH4NO3-ADD discrepancy of the two Gibbs energies, the cross


(METASTABLE
PHASES,V, V* AND II) point of the Gibbs energy curves of phase II (de-
(in J/(mol・deg K)) rived from Table 8) and phase IV (given in Table
7) appears somewhat lower than the observed limit
of the transition point of the complete consistency.
Considering the wide temperature range of the
measurement for the metastable V-V *-II process,
however, the inconsistency appeared on the
transition point of IV-II is unavoidable due to,
the experimental error of heat capacity measure-
ment, and the assumption used for the heat capacity
estimation.

Discussion

(1) Specific Heat Values in Normal Re-


gions. a) Phase V. In Series 3 and 5 of Table
2, the experimental heat capacities of phase V at
the higher temperatures show slight systematic
disagreements between the two series. In Series.
3,the sample was cooled down to 15°K prior to

the measurement, while in Series 5, the sample was.


cooled down to 130°K. Stephenson et al.5)have

reported also a similar effect in the determinationn


of the heat of transition V-IV. They found it was
necessary to cool the sample below 120°K in order

to get the correct values, the pre-cooling to 224°K

giving the lower values by several percents. They


considered that cooling to 224°K was insufficient

to accomplish a complete conversion of phase IV


to phase V.
b) Phase IV. Comparing the data of phase IV
in Series 1 and 6 of Table 2 with each other, the
similar situation found in phase V will also hold for
the discrepancy between them, i. e., the sample
measured in Series I was a virgin sample of phase
IV, whereas the sample measured in Series 6 seems
to contain a small amount of phase III, the heat
capacities of which are much lower than those of
phase IV.
The measurements of the heat capacity of phase-
The relationships in thermodynamic properties
III were carried out in three series of different
between the stable and metastable transition re-
thermal history, and the results were in accordance
gions are presented graphically in Figs. 7 and 8, with each other, showing no evidence of the in-
respectively.
complete conversion of phase IV to phase III.
In Fig. 8, however, the entropy and the enthalpy
As a cause for the persistence of phase III below-
of phase V at 256.2°K are both taken to be zero
the transition point of IV-III, it is reported by
as a standard point in order to show details of the
Whetston19) that such an impurity as potassium
mutual relations among the Gibbs energies.
nitrate is effective for a dry sample.
The Gibbs energy curves of phases IV and II
If a dissolution-recrystallization mechanism20)
derived from Table 7 cross at 324.13°K which is
of transition IV-III causes inhomogeneity due to
compatible with the observed transition point of
the segregation of the potassium impurity in the
IV-II. Furthermore, the heats of transition for
sample, it will be probable that a small impure-
IV-II and III-II at 324.13°K are 1856 and 582
part in the sample retains phase III even below
J/mol from Fig. 7, which are also consistent with the 305.38°K.
observed ones.
At 324°K, the values of entropy, enthalpy and

Gibbs energy function of phase II in Table 7 are 19) J. Whetstone, Acta Cryst., 7, 697 (1954).
20) R. N. Brown and A. C. McLaren, Proc. Roy..
larger than those in Table 8 by -0.012, 0.02 and Soc. (London), A266, 329 (1962).
August, 1967] Heat Capacities and Thesmodynamic Properties of Ammonium Nitrate Crystal 1843

Fig. 8. Gibbs energy vs. temperature curves of NH4NO3. In this figure, the Gibbs energies
are standardized at the state o」phase V at 256.2°K, as follows;

G=H°-H°256.2(V)-T(S°-S°256.2(V))

--- Metastable phase V, V* or II in NH4NO3-add.

As there is no description about the sample they heat curve at around 156°K(-117℃)found by

used, we cannot understand the reason why the us may be looked upon as a corresponding one.
heat capacity data obtained by Eichenauer and The slightness in its effect may be consisting with
Liebscher12) are significantly lower than Ste- the facts that there is no noticeable change in the
phenson's and our data for phases IV and V in its second moment of broad line NMR spectra by
higher temperature region. But, it may be one of Goodwin and Whetstone,22) Gutowsky et al.23)
the possibilities that their sample (V) contains some and by Richards et al.24) The only inconsistent
amount of remained phase III or IV. report against this view is the transition point at
around -160℃ by _Taffrey.9) But there is no
(2) Possibility of the Existence of Phase
VII. As is already described there is a number description about the specific heat data, so we can
of papers which prove the existence of the seventh not check the reliability of his measurement.
phase. But the evidences are not always so definite Anyhow, it seems much desirable to check this
and some of them are contradicting with each other. transition point by using some other techniques.
The transition point at about -60℃ reported by (3) Nature of Phase Transition between
Crenshaw and Ritter,4) and by Jaffrey9) was not V, V* and II. The transition between V and II,
detected by the specific heat measurements by which really consists of two steps, shows some-
Stephenson et al., Eichenauer et al. and by us. what different behaviors from the others. As is
The only seemingly supporting result is the infrared shown in Figs. 4 and 5, the specific heat curve
study by Theoret and Sandorfy.11) They suggest, shows a conspicuous maximum and the Gibbs
however, that this transition point is not corre- energy curve shows no intersection with each
sponding to the transition between IV and VII. other. These facts indicate that this phase transi-
Vol'fkovich et al.,10) who first substantiated tion is of higher order nature. Amoros et al.25)
existence of phase VII, observed this phase at have carried out the complete crystal structure
about -170℃, but they did not report the transi- analysis of phase V at lower temperature which
tion point between V and VII. Judging from these is given in Fig. 9. According to their result,
findings the true transition point between V and the molecular packing quite resembles that of
VII will lie between -60℃ and -170℃. This phase 1113,16)which is shown in Fig. 10. They
expectation seems to be compatible with the ob- suggested that the phase II is a kind of vibra-
servation by Newman and Halford.21) They con-
cluded the significant change in the infrared spectra 22) T. H. Goodwin and J. Whetstone, J. Chem.
Soc., 1947, 1455.
must occur at temperature intermediate between 23) H. S. Gutowsky and G. E. Pake, J. Chem. Phys.
-65 and -160℃ . Taking these consistent ob- 16, 1164 (1948).
servations into account, the small hump in specific 24) R. E. Richards and T. Schaefer, Trans. Faraday
Soc., 57, 210 (1961).
25) J. L. Amoros Portoles, F. Arrese and M. L.
21) R. Newman and R. S. Halford, J. Chem. Phys., Canut, Z. Krist., 117, 91 (1962).
18, 1291 (1950). 26) Y. Shinnaka, J. Phys. Soc. Japan, 11, 393 (1956).
1844 M. NAGATANI, T. SEIYAMA, M. SAKIYAMA, H. SUGA and S. SEKI [Vol. 40, No. 8

some reflections which correspond to (012), (124),


(034) and (142) lines of phase V vanish, and in the
pattern of phase II, furthermore (122), (014) and
(032) lines assigned in regard to phase V disappear.
According to the structure of phase V,25)these re-
flections are attributed only to the hydrogen atoms.
Although there is no detailed information about the
state of nitrate ions in phase V*, the X-ray pattern
of phase V* will assure that the lattice constantst
and the positions of nitrogens are different only
slightly among phases II, V* and V. Considering
the total amount of entropy of transition,⊿S=11

(J/moi・degK)=R ln 4, this transition V-II may


be a result of a kind of combined phenomenon
of orientational order-disorder of both ammonium
Fig.9. A projection on the(001)plane of the
and nitrate groups.
crystal structure of phase V of NH4NO3.25)
P42(C34);Z=8;a=7.98, c=9.78 A at-150℃

Summary

The heat capacities of ammonium nitrate were


determined experimentally between 15 and 410°K

for the stable and the metastable phases. There


are four stable transitions of first-order in this in-
terval, V-IV, IV-III, III-II, and II-I.
Metastable transition IV-II of first-order type
at 324.0°K and metastable one V-II were also

realized. Transition V-II occurs in two steps


via metastable phase V*. The transition points
are found to be 315.2°K for V-V*and 318.2°K

for V*-II, and these transitions are of higher-


order type. The heat capacity of phase V* is
276J/(mol・degK)at 317.2°K.
Fig. 10. A projection on the (001) plane of the
crystal structure of phase II of NH4NO3.13,26) The heats of transition are found to be 473,
P4bm(C24v);Z=2; a=5.74, c=4.95 A 1700, 1350, 4435 and 1860 J/mol for transitions
V-IV, IV-III, III-II, II-I and IV-II, respec-
tional modification of phase V and will undergo tively. The heat of transition and the en-
the direct transformation from phase V under high tropy change were estimated for transition V-II
pressure. One of the present authors, however, as 3254 J/mol, and 10.92)/(mol, degK), using an
realized this transformation at ordinary pressure assumed normal heat capacity. Transition V-II
through the intermediate phase V* by utilizing the is considered to occur as a combined phenomenon
effect of surface active agent.14) We should like of orientational order-disorder of two ions, am-
to point out the features of this phase transition. monium and nitrate groups. The thermodynamic
As is shown in Fig. 4 and Fig. 6 this transition functions have also been derived from the heat
takes place very gradually over a wide range of capacity data. The values at 298.15°K are, in

temperature with no latent heat. Although the units of J/(mol・degK):Cp°=139.1, S°=150.81,

direct transition between phases V and II needs (H°-H0°)/T=79.36 and -(G°-H0°)/T=71.45.

no displacement of the each center of gravity of am-


monium and nitrate ions, this takes place by two The authors wish to express their sincere thanks
steps via phase V*. X-Ray diffraction pattern to Mr. A. Kimoto and Mr. N. Onodera for their
obtained for the powder sample of phase V* is assistances in the experimental work.
very similar to that of phase V as well as to phase
11.14) There may be pointed out some small dif-
† In Fig. 9, the pseudo unit cell (a'→a/√2,
ferences as follows; in the pattern of phase V*, c'→2/c) is illustrated by dotted line.

You might also like